Sov is predicted to be composed of 2499 amino acids; however, information about Sov is very limited. In the present report, we characterize the Sov protein and explore the role of Sov in gingipain secretion

by immunochemical and deletion studies. Strains and plasmids are listed in Table 1. Escherichia coli ER2566 (New England Biolabs) was grown in Luria–Bertani broth. Porphyromonas gingivalis was cultured anaerobically (10% CO2, 10% H2, and 80% N2) at 37 °C in BHIHM [brain heart infusion (Becton Dickinson) supplemented with hemin (7.67 μM) and menadione (2.91 μM)]. Before P. gingivalis cell cultures were used in experiments, the turbidity was adjusted to an OD600 nm of 2.0 using a SmartSpec Plus spectrophotometer (Bio-Rad). Ampicillin (100 μg mL−1) and erythromycin (5 μg mL−1) were added to the medium when needed. PCR was performed with Vent DNA polymerase Buparlisib in vitro (New England Biolabs). this website A 0.5-kbp 5′-terminal region of sov was amplified by PCR with primers 5′-CCGGTACCCATATGTCCGTACCTGCCCGGACTGCC-3′ (italics: NdeI site) and 5′-ACGATATTGCGAGTCTGTGTATTGTCG-3′ and then digested with NdeI and NcoI (in the sov). A 0.3-kbp 3′-terminal region of sov was amplified with primers 5′-GAGCAGCACATCACGAATCCGGAG-3′ and 5′-AATCTAGACCCGGGCAGCTGCGTCAGATTGAAACG-3′ (italics: SmaI site) and then digested with NcoI (in the sov) and SmaI. These PCR fragments were

cloned into the NdeI–PstI sites of pTYB2 with an annealed-oligonucleotide linker (5′-CATCACCATCACCATCACTAGTCTAGAGTCGACCTGCA-3′/5′-GGTCGACTCTAGACTAGTGATGGTGATGGTGATG-3′) to create pKS32. To construct pKS33, a 1.3-kbp sov fragment was amplified with 5′-AAGGTACCATGGGGGCTAAGAGCAATGCAA-3′

(italics: NcoI site) and 5′-AATCTAGACAATACAGGATCGCCAAACGCA-3′ (italics: XbaI site), digested with NcoI and XbaI, and ligated to the NcoI–NheI sites of pTYXB-His (Ishiguro et al., 2009). Similarly, a 1.3-kbp sov fragment was amplified with 5′-AAGGTACCATGGCGAAAAAGTACTGCTTCC-3′ (italics: NcoI site) and 5′-AATCTAGACTGTTTCGGTCGTGCTCCGGCA-3′ (italics: XbaI site), digested with NcoI and XbaI, and ligated to the NcoI–NheI sites of pTYXB-His Immune system to create pKS34. Likewise, the kgp gene was amplified with 5′-CTTCACCATGGATGTTTATACAGATCATGGCGAC-3′ (italics: NcoI site) and 5′-TCTCTAGAACGTACATCGTTTGCAGGTTCGATCGT-3′ (italics: XbaI site), digested with NcoI and XbaI, and ligated to the NcoI–NheI sites of pTYXB-His to construct pKS35. ER2566(pKS32), ER2566(pKS33), ER2566(pKS34), and ER2566(pKS35) were grown in Luria–Bertani broth supplemented with isopropyl-β-d-thiogalactopyranoside (0.3–0.5 mM). Cells were harvested, washed, suspended in 30 mM Tris-HCl (pH 8.0) supplemented with Triton X-100 (2%), sonicated (Ultrasonic generator US-150 with tip #7; Nihonseiki, Japan), and ultracentrifuged (110 000 g for 30 min at 4 °C) to remove the supernatant.

Only 10% of the overall discontinuations observed were because of failure; the short follow-up time might have limited the observation of treatment modification due to failure not occurring as a consequence of intolerance/toxicity or poor adherence. The fact that the reason for discontinuation was determined by the clinician and, as such, was a subjective measure might be seen as a limitation.

However, it was the objective of our analysis to use the clinical perception of the main reason for discontinuation to define the study endpoints. Nevertheless, when we defined discontinuation because of failure on the basis of a viral load >500 copies/mL, or an increase in CD4 cell count IBET762 of <10% from a patient's pre-therapy value or the occurrence of an AIDS-defining illness, the analysis produced results that were very similar to those of the main analysis. Not surprisingly, we found that patients who started therapy with a nonconventional regimen (‘other regimen’) were more likely to

have treatment discontinuation for any reason and for each specific reason than those starting with a standard combination. In conclusion, it seems important to evaluate reason-specific trends in the incidence of discontinuation in order to better understand the determinants of changes over time. The incidence of discontinuation because of intolerance/toxicity has declined over time, Small Molecule Compound Library while simplification strategies have become more frequent in recent years. Despite the fact that drug tolerability has improved and currently available regimens have a reduced pill burden, intolerance/toxicity remains the major cause of drug discontinuation. As reported in our previous study, we confirm that women and HCV-coinfected patients in our cohort are at higher risk of discontinuing HAART. The ICoNA Foundation Study is supported by unrestricted educational grants from Abbott, Boehringer Ingelheim, Bristol-Myers Squibb, Gilead, GSK, Pfizer and Janssen-Cilag. Governing body M. Moroni (Chair), G. Carosi, R. Cauda, F. Chiodo, A. d’Arminio Monforte, G. Di Perri, M. Galli, R. Iardino, G. Ippolito,

A. Lazzarin, F. Mazzotta, R. Panebianco, G. Pastore and C. F. Perno. Steering committee A. Ammassari, A. Antinori, C. Arici, Clomifene C. Balotta, P. Bonfanti, M. R. Capobianchi, A. Castagna, F. Ceccherini-Silberstein, A. Cozzi-Lepri, A. d’Arminio Monforte, A. De Luca, C. Gervasoni, E. Girardi, S. Lo Caputo, F. Maggiolo, R. Murri, C. Mussini, M. Puoti and C. Torti. Participating physicians and centres Italy: M. Montroni, G. Scalise, A. Costantini, A. Riva (Ancona); U. Tirelli, F. Martellotta (Aviano-PN); G. Pastore, N. Ladisa (Bari); F. Suter, F. Maggiolo (Bergamo); F. Chiodo, G. Verucchi, C. Fiorini (Bologna); G. Carosi, G. Cristini, C. Torti, C. Minardi, D. Bertelli (Brescia); T. Quirino (Busto Arsizio); P. E. Manconi, P. Piano (Cagliari); E. Pizzigallo, M.

Only those patients with diagnostic results contribute data for virologic and immunologic analysis, therefore, missing baseline CD4 cell counts or HIV RNA data could have introduced bias into our model estimates. As we are unable to test for any potential bias, this should be taken into account

when interpreting the results of analyses. Patients being VL tested may be retained on failing regimens when second-line therapies are not available. Alternatively, clinicians may not expend scarce resources on diagnostically monitoring patients who are failing clinically and for whom no viable treatment options exist. Consequently, we may be either under- EPZ5676 order or overestimating the proportion of patients who were virologically suppressed. We did not distinguish

between AIDS-related and non-AIDS-related deaths, possibly leading to an overestimation of the number of patients having clinical progression. Patient socio-economic and adherence to therapy data were unavailable. Timely access to CD4 and VL results is crucial for monitoring the efficacy of ARV treatment. These staging data are frequently unavailable in resource-limited settings, and their lack compromises the generalizability of published results and trends. Our analyses included 70% of TAHOD enrollees in disease progression analyses, and 75% (80%) of sites reported that TAHOD patients’ access to VL (CD4) testing did not Entinostat chemical structure differ to that routinely available in their respective countries. Consequently, our estimates of diagnostic resource allocation should be fairly representative of the Asia-Pacific region. However, TAHOD sites are self-selected and patients may differ from other HIV-infected patients within a specific country. Still, our findings highlight challenges for less resourced sites in the region

and potential negative effects on patient outcomes. The from United Nations General Assembly report for the sixty-second session stated that 3 million people from low-income and middle-income countries had access to ARVs in 2007 and that coverage had increased to approximately 30% of those in need [30]. Despite the importance of surrogate laboratory markers in evaluating ARV treatment efficacy, estimates of the availability of diagnostic testing lagged behind treatment access at between 3 and 6% [13]. While recent modelling of HIV infection suggests modest benefits to patient survival from VL monitoring [31], our results show that low levels of site VL testing are associated with poorer treatment outcomes. Further, lack of VL testing increases the risk of patients being maintained on failing regimens and developing highly resistant HIV which may be transmitted to other individuals [32,33].

burnetii BKM120 strain

(RSA493) Nine Mile phase I genomic sequence, which revealed a set of genes with significant homology to the Dot/Icm type IVB secretion system (T4BSS) of Legionella pneumophila. In L. pneumophila, the T4BSS system consists of 26 ORFs, of which 23 share significant homology with C. burnetii ORFs (Seshadri et al., 2003). Studies show that the L. pneumophila T4BSS is required for intracellular survival, effector protein secretion, and replication within host cells (Marra et al., 1992; Berger & Isberg, 1993; Vogel et al., 1998; Bruggemann et al., 2006; Ninio & Roy, 2007; Kubori et al., 2008; Shin & Roy, 2008), thus playing a vital role in the infectious process of L. pneumophila. Moreover, the genome sequence revealed C. burnetii ORFs containing eukaryotic Ankyrin-binding repeat domains (Pan et al., 2008; Voth et al., 2009). Subsequently, these ORFs were shown to be secreted by L. pneumophila in a T4BSS-dependent manner (Pan et al., 2008; Voth et al., 2009), further implicating

the C. burnetii T4BSS as a significant contributor to cellular pathogenesis, Inhibitor Library datasheet and yet characterization of the T4BSS structure in C. burnetii is lacking. In general, T4SS serve to export virulence factors, which include nucleoprotein complexes and effector proteins, into a host or into the extracellular milieu (Christie & Vogel, 2000; Sexton & Vogel, 2002; Cascales & Christie, 2003). T4SS have been subdivided into two families: (1) the VirB/D4 (T4ASS) and (2) the Dot/Icm (T4BSS) systems (Christie & Vogel, 2000).

The T4ASS of Agrobacterium tumefaciens directly injects effector molecules into adjacent cells (Christie Pyruvate dehydrogenase & Vogel, 2000) as well as into the extracellular environment (Dillard & Seifert, 2001; Hofreuter et al., 2001). Interestingly, VirB8, part of the core complex, was reported to localize at the pole of A. tumefaciens cells (Kumar et al., 2000), and the bacterium attaches to host plant cells perpendicular to the bacterial poles (Matthysse, 1987). In L. pneumophila, the T4BSS is essential for cellular pathogenesis via secretion of effector proteins into a host cell (Sexton & Vogel, 2002; Christie et al., 2005). In L. pneumophila, the T4BSS component, DotF, appears to demonstrate polar localization (Jeong et al., 2006). Virulence factors localize or are dispersed about the pole(s) of a wide range of bacteria, and include alternate secretion systems, effector protein molecules, and surface membrane-associated proteins. Evidence suggests that the T3SS of Shigella flexneri is present at the poles of the bacteria before the secretion of IpaC (Jaumouille et al., 2008). Recently, the Mycobacterium marinum Esx-1 T7SS was shown to secrete Mh3864 at the poles and that a core Esx-1 component, Mh3870, localized preferentially to the poles (Carlsson et al., 2009).

7–11.3) was found [39–40]. Early recognition of acute HCV infection is important as treatment with PEG-IFN and ribavirin (RBV) is more successful in acute when compared with chronic HCV infection. The factors associated with HCV transmission

in MSM would seem to be modifiable and potentially amenable to behaviour change interventions and education. To date there have been no RCTs or intervention studies to reduce transmission of HCV in MSM and this should be an area of research. There is also a need to target interventions to prevent HCV reinfection in MSM in particular when access to the new direct acting antivirals (DAAs) will possibly make treatment more effective and more tolerable. There is evidence of delayed anti-HCV seroconversion in HIV-infected individuals. In one study median time from detection Androgen Receptor Antagonist of HCV RNA to anti-HCV detection was 91 days (range 0–1206 days) with 10% failing to seroconvert after 9 months. A low ALT and low nadir CD4 cell count

were associated with a delayed/null anti-HCV response [41]. If individuals are found to be HCV antibody positive, viral load and genotyping measurement should this website be performed. In keeping with racial differences in the anti-HCV responses to PEG-IFN and RBV, single nucleotide polymorphisms (SNPs) in the vicinity of the IL28B locus on chromosome 19 have been found to be associated with the antiviral response [42–43] and spontaneous clearance of HCV in monoinfected populations [44–45]. The C allele at rs12979860 [46] was associated with a favourable response in patients with chronic genotype 1 HCV/HIV infection but less so in those with genotype 2/3 infection or acute HCV [47]. Although the exact mechanism by which this facilitates response

to exogenous IFN-alpha is yet to be elucidated, there appears to be a favourable influence on early viral kinetics [48]. Whilst the CC genotype is associated with a favourable response to PEG-IFN and ribavirin Selleck Decitabine in patients with genotypes 1 and 4 HCV/HIV infection, other factors including HCV viral load and hepatic fibrosis stage also make significant contributions to SVR [48] and the probability of response to PEG-IFN and RBV may be predicted by using algorithms such as the Prometheus Index [49]. With the advent of DAAs and less reliance on augmentation of the innate immune response by interferon, the influence of IL28B SNPs on treatment response and choice and length of therapy will wane [50]. Screening for HDV and HEV are discussed in Sections 7 and 9. We recommend staging of liver disease should be performed in those with chronic HCV/HIV and HBV/HIV infections (1B). We suggest in patients with chronic hepatitis/HIV infection a non-invasive test as the staging investigation of choice (2B).

, 2011). Thus, membrane active agents at sublethal dose are often found to inhibit biofilm formation and thus reduce infection. Consistent with this idea, we have shown here the inhibitory effect of both the alcohols tested against biofilm formation by M. smegmatis. Given its toxicity to mammalian cells and its broad spectrum of target Dabrafenib sites, exploring selective membrane active agents may provide a platform for future drug designs. We would like to thank Ms Urmita Chatterjee and Prof. N. K. Pal, Department of Microbiology, Institute of Post Graduate Medical Education and Research, for their help. We also thank Prof. Sujay Kumar Dasgupta, Bose Institute, Kolkata, for providing the

strain. K.M. is supported by a University Research Fellowship provided by the University of Calcutta. P.T. is supported by CSIR-SRF, Government of India. The AFM facility was made available at the central instrumental facility under DBT-IPLS programme at the University of Calcutta. “
“The purpose of this study was to investigate a three-species in vitro biofilm with peri-implantitis-related bacteria for its variability

and metabolic activity. Streptococcus sanguinis, Fusobacterium nucleatum, and Porphyromonas gingivalis were suspended in simulated body fluid containing MS-275 solubility dmso 0.2% glucose to form biofilms on polished, protein-coated implant-grade titanium disks over 72 h using a flow chamber system. Thereafter, biofilm-coated disks were characterized by scanning electron microscopy and fluorescence in situ hybridization/confocal laser scanning microscopy. To assess metabolic activity within the biofilms, their heat flow was recorded for 480 h at 37 °C by IMC. The microscopic methods revealed that the total number of bacteria in the biofilms varied slightly among specimens (2.59 × 104 ± 0.67 × 104 cells mm−2), whereas all three species were found constantly with unchanged proportions (S. sanguinis 41.3 ± 4.8%, F. nucleatum 17.7 ± 2.1%, and

P. gingivalis 41.0 ± 4.9%). IMC these revealed minor differences in time-to-peak heat flow (20.6 ± 4.5 h), a trend consistent with the small variation in bacterial species proportions as shown by microscopy. Peak heat flow (35.8 ± 42.6 μW), mean heat flow (13.1 ± 22.0 μW), and total heat over 480 h (23.5 ± 37.2 J) showed very high variation. These IMC results may be attributed to differences in the initial cell counts and relative proportions of the three species, their distribution and embedment in exopolysaccharide matrix on the test specimens. The present results provide new insights into variability and dynamics of biofilms on titanium disks, aspects that should be explored in future studies of dental surfaces. Biofilms can be described as communities of microbiota with associated extracellular polymeric matrix on a substrate.

8% to the model. To evaluate the effect of HCV and liver fibrosis parameters in the absence of the effects of ART, we also analysed specifically the subgroup of 110 patients who did not receive ART. The parameters predictive of higher this website CD4 cell count were higher nadir CD4 cell

count (P<0.0001), which by itself explained 83.1% of the total variability in current CD4 cell count, and older age (P=0.006). The remaining parameters, including HCV- and liver fibrosis-related parameters, did not reach the P<0.05 level, although the annual fibrosis progression index was close to it (P=0.06). The variables independently associated with higher HIV-1 viral load in untreated Ku-0059436 in vitro patients were lower CD4 cell count (P<0.0001), younger age (P=0.008), worse CDC clinical stage (P=0.02) and current IDU (P=0.04), accounting for 36.4% of the variability in viral load. HCV and liver fibrosis parameters were not close

to reaching statistical significance. Apart from the study group with active HCV replication, we also recorded the same data for a group of 200 coinfected patients who had cleared the HCV infection, either spontaneously or as a result of anti-HCV therapy. These patients had similar CD4 cell counts and HIV-1 viral loads as patients who had active replication of HCV in the whole group (P=0.5 and P=0.8, respectively). Similar findings were also obtained in the subgroup of patients who were not receiving ART (CD4 cell count, P=0.5; HIV-1 viral load, P=0.4). Multivariate analysis did not show HCV eradication to be independently associated with CD4 cell count or HIV-1 viral load, either in the whole group (P=0.9 and P=0.3, respectively) Pyruvate dehydrogenase lipoamide kinase isozyme 1 or in the subgroup of patients not receiving ART (P=0.1 and P=0.3, respectively). In our study on a large population of HIV-1-infected patients with active HCV infection, we found that HCV-related variables did not significantly influence the virological and immunological outcomes of HIV-1 infection, after adjusting for other covariates. In contrast, liver fibrosis, as measured using the annual fibrosis progression

index, was independently predictive of CD4 cell count, although its influence was relatively small. A number of studies analysing the effect of HCV infection on clinical and HIV-1 viroimmunological parameters have been published, with contradictory results, mainly attributable to different designs, sample sizes, outcomes evaluated and patients’ characteristics. Regarding clinical outcomes, some studies reported that there was no significant effect of HCV on clinical progression to AIDS or death [1,7,28–33,36,37]. However, despite the absence of differences overall, some studies, not surprisingly, found that morbidity and mortality related to liver damage were more common in HIV-1/HCV-coinfected patients [1,36].

Importantly, the definition of treatment failure (virological, immunological

or clinical) was the strongest predictor of resistance. Immunological and clinical treatment failure were categorised as inclusion criteria because access to plasma HIV-1 RNA and CD4 quantification was irregular during the study period. The finding that immunological and clinical criteria were poor predictors of treatment failure attributable to resistance is important and has relevance for other resource-poor settings where access to VL testing may be limited. Our results are in agreement with recent Selleck MLN0128 data which also show that CD4 cell counts and clinical criteria do not accurately identify patients with virological treatment failure [17–19]. In this study, we estimated that the overall prevalence of resistance-associated mutations was 81% among the investigated Honduran patients,

who were selected on the basis of signs and symptoms of treatment failure. This finding agrees well with results from the United Kingdom (80% resistance) [10], United States (76% resistance) [11], and France (88% resistance) [12], in spite of the fact that the cART conditions in these countries are very different from those in Honduras. It is somewhat surprising that 19% of the 138 studied patients did not appear to harbour a resistant virus even though they had been selected as experiencing treatment failure. The absence of resistance in 19% of the patients indicates that adherence in these patients may have been AZD8055 ic50 too low to drive the development of resistance. However, and as discussed above, the significant association of resistance

with type of failure (virological, immunological or clinical) also demonstrates that it is difficult DNA Damage inhibitor to monitor HIV therapy without regular access to plasma HIV-1 quantification. Thus, patients with adequate adherence and low plasma HIV RNA levels may incorrectly have been perceived to have immunological or clinical treatment failure. We observed that the presence of genotypic resistance was positively correlated with years on therapy and the number of ART regimens used. When this study was carried out, access in Honduras to boosted PIs was very limited. The broad resistance to NRTIs, NNRTIs and unboosted PIs indicates that many of our study subjects were in need of salvage therapy with boosted PIs as well as entry and integrase inhibitors [3]. Improved access to these drugs is urgently needed for these and other heavily treatment-experienced Honduran patients. M184V and K103N were the most prevalent NRTI and NNRTI mutations in our study, at 62% and 30%, respectively. M184V causes high-level (>100-fold) resistance to lamivudine/emtricitabine and emerges rapidly in patients who receive lamivudine monotherapy [20]. It is also the first mutation to develop in patients receiving partially suppressive triple combination therapy including lamivudine [21–23].

R.L., Buenos Aires, Argentina). Seventeen Argentinean F. poae isolates from different regions and

hosts selected at random were analysed by HPLC/FD to test NIV/DON production (Table 1). Fusarium poae isolates were cultured in Erlenmeyer flasks (250 mL) containing 25 g of long-grain rice. Ten mL of distilled water was added before autoclaving for 30 min at 121 °C, twice. Each flask was inoculated with a 3-mm diameter agar disc taken from the margin of a colony grown on SNA (Nirenberg, 1976) at 25 °C for 7 days. Flasks were shaken once a day by hand for 1 week. These cultures were incubated for 28 days at 25 °C in the dark. At the end of the incubation period, the contents of the flask 17-AAG in vitro were dried at 50 °C for 24 h and then stored at −20 °C until being analysed for toxin. Toxin extraction and clean-up were carried out using a modified version of that originally reported

by Cooney et al. (2001). For the detection of NIV and DON, the analysis was performed using the conditions described by Barros et al. (2008). The dried residue was dissolved in 400 μL of methanol/water (5 : 95), homogenized in a vortex mixer and injected into the HPLC system by full-loop injection technique (Hewlett Packard model 1100 pump, Palo Alto, CA and Rheodyne manual injector with a 50 μL loop; Rheodyne, Cotati, CA). The HPLC system consisted of a Hewlett Packard model 1100 pump connected to a Hewlett Packard 1100 Series variable wavelength detector and a data module Hewlett Packard Kayak XA (HP ChemStation Rev. A.06.01). selleck Montelukast Sodium Chromatographic separations were performed on a Luna™ C18 reversed-phase column (100 × 4.6 mm, 5 μM particle size) connected to a guard column SecurityGuard™ (4 × 3.0 mm) filled with the same phase. The mobile phase consisted of methanol/water (12 : 88),

at a flow rate of 1.5 mL min−1. The detector was set at 220 nm with an attenuation of 0.01 AUFS. Quantification was relative to external standards of DON and NIV (Sigma-Aldrich Co., St Louis, MO) from 1 to 4 μg mL−1 in methanol/water (5 : 95). The detection limit was 5 ng g−1 for each toxin. Fusarium poae is recognized as a more prominent member of the FHB complex (Yli-Mattila et al., 2008; Kulik & Jestoi, 2009; Stenglein, 2009). Several researchers have developed specific primer pairs for PCR assays, to have a rapid, inexpensive and relative simple technique to identify F. poae isolates of cereal samples (Parry & Nicholson, 1996; Kulik, 2008; Yli-Mattila et al., 2008). Fusarium poae isolates used in our study were found to be positive based on the specific primer pair developed by Parry & Nicholson (1996). Seventeen Argentinean isolates were analysed by HPLC/FD for production of trichothecenes and only NIV was detected (0.3–8.7 μg g−1; Table 1). This was in agreement with other studies (Vogelgsang et al., 2008a ,b; Yli-Mattila et al.

All patients were required to have a suppressed viral load, defined as a viral load ≤500 copies/mL, at baseline. Patients were excluded if there was no viral load meas-urement in the 6 months after Sunitinib nmr baseline. Virological failure was

defined as a viral load >500 copies/mL measured at least 4 months after baseline. Patient follow-up was measured from baseline to date of virological failure or date of last viral load measurement. Poisson regression analysis was used to identify viral load response prior to baseline associated with virological failure after starting new ARVs. Potential explanatory variables included age, gender, year of starting cART, ARV exposure status at cART initiation (ARV-naïve or ARV-experienced), risk group, ethnicity, region of Europe, baseline

CD4 cell count, CD4 nadir, peak viral load, previous AIDS diagnosis, time on cART, current Selleckchem Y-27632 treatment regimen, number of previous treatment regimens, time spent on cART prior to baseline, number of ARVs to which the patient was previously exposed and the reason reported for starting the new ARV. In addition to the traditional explanatory variables investigated above, variables that summarized the history of viral suppression after cART initiation prior to baseline were investigated. The variables used to summarize the history of viral suppression after cART initiation were as follows. 1 Months to initial suppression (HIV RNA ≤500 copies/mL) after starting cART. Viral suppression was

defined as a measurement of HIV RNA ≤500 copies/mL. Viral rebound was defined as a viral load >500 copies/mL measured after a period of suppression prior to the regimen change. For variable 5, any period of time when the patient was off cART and the first 4 months after starting a new cART regimen were excluded. Thus, only periods during which the patient was on cART and should have been virally suppressed were included. Any variable that was significant at the 10% level in the univariate model was then included in a multivariate model. The sensitivity analysis considered confirmed virological failure after baseline (i.e. two consecutive viral load measurements above 500 copies/mL) and, in the subgroup of patients who had viral load measured using an assay with a lower limit filipin of detection of 50 copies/mL, virological failure after baseline defined as a viral load above 50 copies/mL. Analyses were also repeated taking account of HIV drug resistance at baseline in the subset of patients with resistance data, using genotypic sensitivity scores (GSS) calculated using the rega algorithm, version 7.1 [29]. A total of 1827 patients (67%) were included in the analysis. Table 1 describes the characteristics of the patients included in the analysis. Eight hundred and seventy-eight patients (48%) were treatment naïve at cART initiation.